My first step is getting a brand new computer all together! Seeing this cool stuff is some good motivation lol. I’ve been exclusively on my phone and tablet the past 2 years because my computer is too old and slow! I’m gonna get a good one👍
Why when I put prompts to create an imagine in Stable Difussion, it takes time to create but when it reaches 100% , it just dissapear and I don't even see it
I'm having trouble downloading it it says it couldn't find a version that satisfies the requirement and no matching distribution was found for torch. What should I do?
stderr: ERROR: Could not find a version that satisfies the requirement torch==1.12.1+cu113 (from versions: none) ERROR: No matching distribution found for torch==1.12.1+cu113
I struggled for a while because of a lot of misinformation out there. In order to get git working I had to add the bin and cmd folder from the installation folder to the environment variables/paths in windows. Hope this helps someone.
Hey, can you make video about how to make public link? I want to run stable diffusion on PC and use it on my phone but i don't know where set that "share". I tried in comand args --share and --listen but it then says something is wrong with my internet and it can't create public link. Older version of SD worked fine but now there is issue Also thanks for video, great edit, calm voice and everything fast explained! You deserve more subs, keep going
My system kept crashing a few seconds in to generation (way before cuda and ram reached max) (2080ti 10700k) Had to undervolt the GPU and that seems to have fixed it.
I got an error while trying to install it- "Torch is not able to use GPU" It then follows up with "CUDA initialization failed, you might not have a CUDA GPU." Any ideas for a possible fix?
ill just wait for some preinstalled buy version, too many problems with it, great video, very informative, but redone it a couple times and the ip link says unable to connect. maybe something good will be released soon for a download ready to go
AT 5:54 - "copy as path" does NOT appear in my Windows 10 (fully updated) right-click context menu (but I figured it out). Also, when I add that "git pull", I get the error: fatal: not a git repository (or any of the parent directories): .git
You don't need git, just Python 3.10.6 and Microsoft Windows 10 SDK. You get that from Microsoft visual studio installer. The SDK itself is free, don't download everything (it's too big) just download Windows 10 SDK from there. That thing would take care for most of the errors (if not all of them).
Hi I am getting "fatal detected dubious ownership in repository" error everytime don't know how to solve that. Can you help me solve that please. Thank you for this video it's really awesome.
on amd it only works on linux and i have a dual pc setup and my rtx 3070 its giving me the torch error and i deleted the files and nothing i wish i could of used my 6900xt instead
Probably not, even if the model could fit in the (virtual) video RAM, performance would be so low that it'll really struggle. Also, Intel UHD graphics covers all integrated graphics from Intel, this really needs either nVidia, AMD or even Intel dedicated graphics cards to work right.
stderr: ERROR: Could not find a version that satisfies the requirement torch==1.12.1+cu113 (from versions: none) ERROR: No matching distribution found for torch==1.12.1+cu113 help please, i try to delete venv and run again, but same problem, please if you know
You can drag/drop, or select images on the upscaling tab. Then you can mix and choose the upscaling methods - as well as bulk upscale from a folder - Yes.
modules.devices.NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.
Amazing guide. Thanks so much for spelling it out in a a way that's really helpful for an old noob such as myself. Can't wait to get creating
My first step is getting a brand new computer all together! Seeing this cool stuff is some good motivation lol. I’ve been exclusively on my phone and tablet the past 2 years because my computer is too old and slow! I’m gonna get a good one👍
Thanks a bunch. Finally installed Automatic 1111
So when you close it out how do you open it back up?
This was really good! Thank you so much! Lots of detail but also nicely paced. Thank you for including what to do when things go wrong!
cheers
Why when I put prompts to create an imagine in Stable Difussion, it takes time to create but when it reaches 100% , it just dissapear and I don't even see it
I appreciate updates on Stable Diffusion/AI image generation ❤
I'm having trouble downloading it it says it couldn't find a version that satisfies the requirement and no matching distribution was found for torch. What should I do?
how do you get ur diffusion dark colored page? Mine is completly white
It only says "already up to date" and nothing happens.
Can please somebody help me with that?
stderr: ERROR: Could not find a version that satisfies the requirement torch==1.12.1+cu113 (from versions: none)
ERROR: No matching distribution found for torch==1.12.1+cu113
exact thing happened to me and i dont know how to fix it
@@englishbulldog651 Uninstall python, instal an older version 3.10.6.
@@ChosenBlood1 not all heroes wear capes!
Excellent video!
Thanks , i think the wiki change, the models are not available now
I struggled for a while because of a lot of misinformation out there. In order to get git working I had to add the bin and cmd folder from the installation folder to the environment variables/paths in windows. Hope this helps someone.
How?
how
it finished downloading but didnt get me a link can you help?
stable diffusion runtimeerror couldn't install gfpgan
Hey, can you make video about how to make public link? I want to run stable diffusion on PC and use it on my phone but i don't know where set that "share". I tried in comand args --share and --listen but it then says something is wrong with my internet and it can't create public link. Older version of SD worked fine but now there is issue
Also thanks for video, great edit, calm voice and everything fast explained! You deserve more subs, keep going
Very useful, thx!
Getting an error when running webui-user and it cant be reached. “Interrupted with signal 2…”
I enter the URL, but it doesn't work. How can I check what's wrong?
when i do the command prompt it doesn't it downloads all the files on the website files.
when I try to generate an image the console appears to be doing nothing.
Thanks, I was wondering why I was stuck on installing rich and torchvision, guess it just takes long cause of ram
Couldn't install torch... dont work delete venv and temp
My system kept crashing a few seconds in to generation (way before cuda and ram reached max) (2080ti 10700k) Had to undervolt the GPU and that seems to have fixed it.
I got an error while trying to install it- "Torch is not able to use GPU" It then follows up with "CUDA initialization failed, you might not have a CUDA GPU." Any ideas for a possible fix?
super helpful, although, I didnt get any error messages nor did a link generate. Do you know what I do in this instance?
ill just wait for some preinstalled buy version, too many problems with it, great video, very informative, but redone it a couple times and the ip link says unable to connect. maybe something good will be released soon for a download ready to go
Automatic Installation now there are only 4 points not 6
'git' is not recognized as an internal or external command,
operable program or batch file..
i watched and read tones of tutorials and nothing is working..
Hi there, I download and install the staple diffusion but it doesent go beyond " Installing torch and torchvision ". what to do? HELP!!?
Same here for 2 hrs
@@rave6893 What to do now?
same problem...
In case this helps, you need Python 3.10.6. This video had a mistake of showing 3.11 which will not work.
@@Avenger222 I did install Python 3.10.6.
AT 5:54 - "copy as path" does NOT appear in my Windows 10 (fully updated) right-click context menu (but I figured it out). Also, when I add that "git pull", I get the error: fatal: not a git repository (or any of the parent directories): .git
You don't need git, just Python 3.10.6 and Microsoft Windows 10 SDK. You get that from Microsoft visual studio installer. The SDK itself is free, don't download everything (it's too big) just download Windows 10 SDK from there. That thing would take care for most of the errors (if not all of them).
for the copy as path u just have to hold shift while you right click
doesnt work
Hi I am getting "fatal detected dubious ownership in repository" error everytime don't know how to solve that. Can you help me solve that please. Thank you for this video it's really awesome.
Install on your c: drive and make sure the drive is in ntfs system format.
Trying to get this to work on a AMD RX-580…
on amd it only works on linux and i have a dual pc setup and my rtx 3070 its giving me the torch error and i deleted the files and nothing i wish i could of used my 6900xt instead
Hi there! Would any of this work with Intel(R) UHD Graphics ?
Probably not, even if the model could fit in the (virtual) video RAM, performance would be so low that it'll really struggle. Also, Intel UHD graphics covers all integrated graphics from Intel, this really needs either nVidia, AMD or even Intel dedicated graphics cards to work right.
help
C:\Users\User\AppData\Local\Programs\Python\Python310\python.exe: No module named pip
Launch unsuccessful...
5:53 there is no copy as path option
if you do this, you just have to add python.exe to the end
Shift right-click worked for me.
You have to hold shift
doesn't work what a fake
wow thats like 1000 steps too many for me to bother with
Hey I got it to work! ty!
Does this work for AMD 7900xtx yet? Specifically RDNA3 on Vulcan?
stderr: ERROR: Could not find a version that satisfies the requirement torch==1.12.1+cu113 (from versions: none)
ERROR: No matching distribution found for torch==1.12.1+cu113 help please, i try to delete venv and run again, but same problem, please if you know
Same here
Check what version of python you are using. I was using 311 and when I got 3.10 instead it fixed the error for me.
@@bigal1093 Check what version of python you are using. I was using 311 and when I got 3.10 instead it fixed the error for me.
You need version 3.10.6 of python, you might be using 3.11
@@AyeDeeHD Yup ... that was it ... thanks!!
thanks guys now it's work !! but i have problem RuntimeError: CUDA out of memory.
Quick question. Can i use this to upscale a photo that isn't ai generated?
You can drag/drop, or select images on the upscaling tab. Then you can mix and choose the upscaling methods - as well as bulk upscale from a folder - Yes.
i have checked the python path several times, copying and pasting, but next step it always says python does not exist at the path location
Make sure "add to path" is ticked on the first page when installing python
@@TroubleChute yeah just did that. Thanks!!!
how do you add torch to start with discrete gpu? its not an exe that i can find
figured that part out but it doesnt run with my 6800m, starting the .bat says torch is not allowed to use this gpu, error x02......long code
@@TroubleChute running, thanks for the great video man
modules.devices.NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.
How can i fix this